Takeaways from the Djed AMA

COTI has hosted a written AMA in their Telegram channel, today.

Lots of questions were about the general workings of Djed, which could have been answered by the paper (https://iohk.io/en/research/library/papers/djeda-formally-verified-crypto-backed-pegged-algorithmic-stablecoin/), the IOHK blogpost (https://iohk.io/en/blog/posts/2021/08/18/djed-implementing-algorithmic-stablecoins-for-proven-price-stability/), or the COTI FAQ (https://medium.com/@cotinetwork/djed-frequently-asked-questions-f636735be76).

Interesting and not answered before for me were these:

They are going to restrict access to “the DJED app”: “Currently, the DJED app is restricted in 30 countries, including the US, Israel, Canada, Albania, and other sanctioned countries. This was a necessary step to be able to open the minting of SHEN to everyone.” https://t.me/COTInetwork/743257

I don’t know how that works technically. In pure doctrine, smart contracts should not be able to be restricted. But since Plutus has this strange off-chain/on-chain architecture, it is probably possible by just restricting access to the website, since interaction with the contract is not possible without the centralised web app.

They are going to stake the reserve: “Therefore, to create another revenue stream for our SHEN holders, we will delegate the ADA in the contract to specific stake pools, and 100% of the rewards received will be transferred to the reserve (increasing the reserve ratio) and later distributed to the SHEN holders.” https://t.me/COTInetwork/743289 “Very important to clarify that the stake pool operators that we will be working with will be screened and very carefully picked according to different parameters to ensure the stability, security, and functionality of the contract is not affected.” https://t.me/COTInetwork/743291

I’m not sure if they really understood, how staking on Cardano works. Functionality of the contract cannot be affected by staking the contract address ever. But it’s nevertheless good to have some due dilligence to ensure that the staking rewards really arrive.

The contract will be open-sourced at some time, but the “When?” in the question was not answered: “According to the above, it makes total sense to make open source the DJED code and that’s precisely what we will do. The research paper is written in the most detailed way possible because we want users not only to use it but also to understand it, and part of that process is seeing the code as well.” https://t.me/COTInetwork/743326

In contrast to an answer given to @ChrisSTR8 by one of their admins previously (https://t.me/COTInetwork/733635), they will implement the Minimal Djed variant from the paper linked above, not the Extended Djed variant: “We are going to implement minimal Djed, but we are exploring the possibility of implementing the Extended Djed in the future” https://t.me/COTInetwork/743329

Interestingly, the additional fee going to COTI (or their token holders) was obviously not discussed with IO: “Another risk that occurred to me today, as I learned about how operational fees are going to be managed is that the sale of ADA for COTI will naturally create some downward price pressure on ADA w.r.t. to COTI. Again, this should be negligible if the volume of these operational fees is small. But it could become non-negligible, if the volume grows.” https://t.me/COTInetwork/743344 The main author of the Djed paper only learned about it today. …

I would still be interested in more details about how the oracle giving the ADA/USD exchange rate to the Djed system works. Which exchanges are taken into account? Who controls the oracle and can change that? What happens if the oracle stops working? https://t.me/COTInetwork/743308 Unfortunately unanswered.

Another unanswered question: “Every transaction needs to know the current Djed and Shen in circulation and the current ADA in the reserve as per the mathematics. So, only one transaction can be created, signed, and processed at a time. How did you solve that bottleneck with a Plutus contract?” https://t.me/COTInetwork/743351

So, I’m still looking forward to the open-source release of their Djed implementation, which hopefully will answer some of that.

EDIT: The transcript of the AMA is now online: https://cotinetwork.medium.com/djed-ama-recap-6c710f70e441


Another question: what is the use of COTI in the context of Djed?

1 Like

That is starting to not sound as “open” and “decentralised” as expected by the Cardano community.

If we want Djed to be a successful decentralised algorithmic stable coin then:

  1. Anyone must be able to mint Shen.
  2. There must be no “Djed app” restrictions.
  3. The code must be open sourced.

I sense an enormous opportunity with Djed. If Coti doesn’t get this sorted as the community expects then someone else needs to.


I don’t know if I understand the question. COTI, the company, is the company implementing it. And part of the fees will be used to buy COTI, the token, to be put in their treasury and benefit COTI, the community.

Other than that, there is no tight connection.

1 Like

There is a crying need for a truly decentralised stable coin. DAI is not it. Dai is backed significantly by centralised assets and has added governance complexity/issues.

Cardano has a chance to own this market. This opportunity is so huge. The slate has been recently wiped clean of competitors. Furthermore, the risk of regulation looms for anything with centralised attack points.

Djed needs to grab this opportunity with both hands. For it to be successful it needs to be set free. We don’t want any centralised limits placed on this thing. That means Coti must not limit any activities related to operation of the protocol. And, that includes the minting of Shen.

This will be like shooting ourselves in the foot if Cardano does not seize this opportunity.

The market cap for a truly decentralised stable coin will be in the trillions. Think about how much Ada needs to back that and what that implies for the price of Ada. Furthermore, the decentralised stable coin market will tend towards one predominant winner because people/businesses/govts will prefer the leader with most liquidity so they don’t move the market with large trades.

1 Like

Sorry. I meant COTI the coin.

I’ve seen some reddit posts that hopefully can explain it for you.
The significance of coti’s existence lies in policy compliance and some financial licenses.

It would be nice if you could link a couple of those posts, to help followers of this thread (including me, not generally a Reddit user) look at that particular angle. I don’t want to say anything bad about COTI but if their product is closed source then a lot more needs to be known about the company in order to trust them.

This explanation is good enough for me and also what I assumed from the beginning when I heard they were behind the new Djed. Yet another reason to bookmark your Reddit links while they still may be current & easier to find, to try to fill out the whole story. :face_with_monocle:

1 Like

You can search for djed in the cardano or coti modules and browse related discussions. Hope this post helps you.


I totally agree with you.

COTI’s Djed might be OK as Cardano’s first stablecoin implementation, but I will never fully trust it. I also think we should move away from USD as the default currency to peg to.

Some thoughts

We all agree that inflation is bad. So why should we peg to USD (or any other fiat)?

Deflation is also bad as nobody would want to borrow in such a currency. And credit is an important part our economies, right?

So I think our community stablecoin should use a mechanism inspired by the Consumer Price Index. Designing such an oracle isn’t easy, but not impossible. There are plenty of online price-feeds for standardized goods and services (online supermarkets etc.). The stablecoin governance can be used to vote which version of the CPI web-crawler to use.

We could call it World Stable Coin (or World Food Token if the CPI is mostly food-based), etc… Because it’s not fiat-based I don’t think financial regulators will care as much as conventional stablecoins. Users of this kind of stablecoin will clearly know its not fiat, and they should naturally be more aware of the risks. I think that that is really what financial regulators want: for investors to be fully informed (TerraUSD holders were clearly not informed).

The CPI oracle should be organized as an oracle pool, and anyone should be able to participate by buying oracle membership tokens. I also think that the oracle membership tokens and the stablecoin governance tokens should be the same token. (Governance holders don’t have an on-chain way of knowing which CPI web-crawler is used, so they would have to vote to punish the whole oracle pool if they refuse to use that version. So as a backup there would need to be multiple oracle pools competing to provide the same information. Oracle pool members can however easily do an on-chain check of any deviation of the other members’ web-crawler, and can punish individual members for not updating, without impacting stability of the oracle pool itself.)

For parallelization I would have the same stablecoin script sit at different staking addresses. Any Djed- or Shen-equivalent can be used in any of the parallel script addresses. I think market effects should eventually balance the equity and reserve-ratios of each parallel script. We would though have to find a way to withdraw staking rewards as equity ‘automatically’.

The nice thing about staking (and really the beauty of PoS), is that regular ADA stakers get ~5% APY, but people who are willing to accept the leveraged exposure to ADA’s price movements, offered by the stablecoin reserve currency, can get ~10% APY for twice the risk (at 100% reserve-ratio). The larger the reserve-ratio, the lower the leveraged staking reward, but also the lower the risk.

Ehm, not all, but I know very well that I seem to be a minority, here.

The consumer price basket based stablecoin is interesting, nevertheless.